IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
On systems without /proc/fd support this avoid the expensive chdir()
logic in non_widelink_open(). open_file_ntcreate() already passes
dirfsp and atname to reopen_from_fsp(), it was just missed in the
conversion.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15549
Reviewed-by: Volker Lendecke <vl@samba.org>
Signed-off-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Mon Jan 22 12:00:56 UTC 2024 on atb-devel-224
(cherry picked from commit 2713023250)
Autobuild-User(v4-18-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-18-test): Mon Jan 29 11:53:56 UTC 2024 on atb-devel-224
Existing caller passes NULL, no change in behaviour. Prepares for
replacing symlink_target_below_conn() in open.c.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15549
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
(cherry picked from commit fc80c72d65)
Moves processing the symlink error response to the caller
filename_convert_dirfsp(). Prepares for using this in
non_widelink_open(), where it will replace symlink_target_below_conn()
with the same functionality.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15549
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
(back-ported from commit 0515dded4d)
This is a more sensible combination of missing Linux specific features:
- O_PATH
- openat2() with RESOLVE_NO_SYMLINKS
- somehow safely reopen an O_PATH file handle
Currently only O_PATH is disabled for these jobs, but that doesn't really match
and know OS.
The following list shows which features are available and used by Samba on a few
OSes:
| O_PATH | RESOLVE_NO_SYMLINKS | Safe reopen | CI covered
--------|----------------|---------------------|----------------------------
| Supported Used | Supported Used | Supported Used |
============================================================================
Linux | + + | + + | + + | +
FreeBSD | + + | + [1] - | + [2] - | -
AIX | - - | - - | - - | +
So by also disabling RESOLVE_NO_SYMLINKS and Safe Reopen, we cover classic UNIX
systems like AIX.
[1] via open() flag O_RESOLVE_BENEATH
[2] via open() flag O_EMPTY_PATH
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15549
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
(cherry picked from commit 62cbe145c7)
This will be used in CI to have a gitlab runner without all modern Linux
features we make use of as part of path processing:
- O_PATH
- openat2() with RESOLVE_NO_SYMLINKS
- somehow safely reopen an O_PATH file handle
That gives what a classix UNIX like AIX or Solaris offers feature wise.
Other OSes support other combinations of those features, but we leave the
exersize of possibly adding more runners supporting those combinations to the
reader.
The following list shows which features are available and used by Samba on a few
OSes:
| O_PATH | RESOLVE_NO_SYMLINKS | Safe reopen | CI covered
--------|----------------|---------------------|----------------------------
| Supported Used | Supported Used | Supported Used |
============================================================================
Linux | + + | + + | + + | +
FreeBSD | + + | + [1] - | + [2] - | -
AIX | - - | - - | - - | +
[1] via open() flag O_RESOLVE_BENEATH
[2] via open() flag O_EMPTY_PATH
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15549
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
(cherry picked from commit 5c2f96442a)
Browsing files or download files from samba server, smbd would check user's
id to decide whether this user could access these files, by lookup user's
information from the password file (e.g. /usr/local/samba/private/smbpasswd).
smbd might goes through startsmbfilepwent(), this api calls [f]chmod() to
make sure the password file has valid permissions 0600.
Consider a scenario: we are doing a read performance benchmark about
downloading a bunch of files (e.g. a thousand files) from a samba server,
monitoring file system i/o activities counters, and expecting that should
be only read operations on file system because this is just downloading, no
uploading is involved. But actually found that still write operations on file
system, because smbd lookup user and always reset 0600 permissions on password
file while access each file, it makes dirty pages (inode modification) in ram,
later triggered a kernel journal daemon to sync dirty pages into back storage
(e.g. ext3 kjournald, or ext4 jbd2).
This looks like not friendly for read performance benchmark if it happened on
an entry-level systems with much less memory and limited computation power,
because dirty pages syncing in the meantime slows down read performance.
This patch adds fstat() before [f]chmod(), it would check whether password
file has valid permissions 0600 or not. If 0600 smbd would bypass [f]chmod()
to avoid making dirty pages on file systems. If not 0600 smbd would warn and
go through [f]chmod() to set valid permissions 0600 to password file as
earlier days.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15555
Signed-off-by: Jones Syue <jonessyue@qnap.com>
Reviewed-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Thu Jan 18 10:28:19 UTC 2024 on atb-devel-224
(cherry picked from commit c82a267b2a)
Autobuild-User(v4-18-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-18-test): Mon Jan 22 10:09:52 UTC 2024 on atb-devel-224
Doing the previous version access checks and semantics at the SMB
layer means we can simplify the shadow_copy2 and remove the kludge.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13688
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Mon Jan 8 16:58:26 UTC 2024 on atb-devel-224
(backported from commit f14a706569)
[slow@samba.org: vfs_shadow_copy2.c: no TALLOC_FREE() in context]
[slow@samba.org: open.c: assign result from calculate_open_access_flags()]
Autobuild-User(v4-18-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-18-test): Tue Jan 16 10:05:29 UTC 2024 on atb-devel-224
Now that check_any_access_fsp() is broadly used consistently to
restrict access for all modifying operations, we can add a check for
previous versions to check_any_access_fsp() and it gets enforced
consistently.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13688
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit fd4e41144a)
Replaces the direct access to fsp->access_mask with a call to
check_any_access_fsp() which allows doing additional checks if needed.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13688
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(backported from commit 02ed99343d)
[slow@samba.org: vfs_acl_common.c: different chown_needed check]
The additional check if fd underlying fd is valid and not -1 should not be done
at this place. I actually would prefer an write to fail with EBADF if this
happens, as it's likely easier to debug why this happened. These days we should
always have a valid fd.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13688
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 995a31c8d4)
smbd_check_access_rights_fsp() requires *all* rights in access_mask to
be granted by the underlying ACL, but the semantics of this function
is supposed to grant access if any one of the rights in
access_requested is allowed.
Fix this by looping over the requested access mask. If
smbd_check_access_rights_fsp() returns sucess, mask will be non-null
and when assigned to access_granted, the subsequent check will pass,
fail otherwise.
I'm not doing an early exit on purpose because a subsequent commit
adds additional security checks that are done in the subsequent code
path common for fsa and non-fsa fsps.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13688
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit bf497819e6)
The semantics of the access check in check_access_fsp() itself is to
allow access if *at least* one or more rights of the rights in
access_mask are allowed. The name check_any_access_fsp() better
reflects this.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13688
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 96b577c380)
Printer file handles went through SMB_VFS_CREATE_FILE() and are network
callable, so it makes sense to set this on them.
This ensures that check_access_fsp() doesn't take the codepath calling
smbd_check_access_rights_fsp(), but just checks the request rights from
fsp->access_mask.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13688
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 76c8fe16bf)
Test more modifying operations are blocked and access masks are correct.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13688
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 537eedfe2a)
Frankly, I can't remember why I added this as part of bug 13688. The
goal of the corresponding test is to verify a write on a read-only
file handle fails. As the file is opened O_RDONLY, the write will fail
anyway and there's no need to inject the error.
To make things worse, having the error injected meant we didn't notice
when the underlying logic of forcing the open to be done with O_RDONLY
was done as O_RDWR, resulting in the write on the handle to succeed.
This happened when we introduced reopen_from_fsp(): the initial
pathref open of a path with a twrp value was correctly detected and
handled by shadow_copy2_openat(). However, when converting the pathref
open to a real one via reopen_from_fsp(), shadow_copy2_openat() only
sees the magic /proc/fd path and has no way of inferring that this was
originating from a prevous version open with a twrp value.
Tl;dr: we can just remove this error injection, it is not needed, the
correct fix is to implement this in the SMB layer which is done in the
subsequent commits.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13688
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 78119edba0)
Thanks to Thierry LARONDE for the fix.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15540
Signed-off-by: Björn Jacke <bjacke@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit b9f32b32e0)
Autobuild-User(v4-18-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-18-test): Mon Jan 8 14:02:46 UTC 2024 on atb-devel-224
Bug: https://bugzilla.samba.org/show_bug.cgi?id=15544
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Jan 2 20:37:01 UTC 2024 on atb-devel-224
Right now we can't traverse a subdirectory in a snapshot which was
deleted in the current set of files.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=15544
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
(cherry picked from commit a0bfadb32b)
To maintain SMB Multichannel, windows client might periodically query with
FSCTL_QUERY_NETWORK_INTERFACE_INFO to get SMB server's network information,
in my case windows server 2022 would do this every 10 minutes (600 seconds).
Consider a scenario: the network information might have changed between
these queries, some become link down, new interface is link up, network
speed is changed, and etc. So far smbd might not aware of these changes and
still report out-of-date network information to windows client, until we
manually send a SIGHUP to smbd in order to trigger load_interfaces():
smbd_sig_hup_handler() > reload_services () > load_interfaces()
This might be a bit inconvenient because it is hard to decide when should
we manually send a SIGHUP to smbd for refreshing network information.
This patch adds load_interfaces() at fsctl_network_iface_info(), while smbd
received FSCTL_QUERY_NETWORK_INTERFACE_INFO would go through this and refresh
local_interfaces, then respond to client with up-to-date network information;
also refresh num_ifaces to make sure interfaces count is consistent.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15547
Signed-off-by: Jones Syue <jonessyue@qnap.com>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Björn Jacke <bjacke@samba.org>
(cherry picked from commit 318fd95d5e)
Autobuild-User(v4-18-test): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(v4-18-test): Fri Jan 5 13:46:39 UTC 2024 on atb-devel-224
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15534
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Thu Dec 21 11:09:30 UTC 2023 on atb-devel-224
(cherry picked from commit 828f3c9912)
Autobuild-User(v4-18-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-18-test): Thu Dec 28 17:09:25 UTC 2023 on atb-devel-224
Modernise debug while touching the code.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523
Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Autobuild-User(master): Stefan Metzmacher <metze@samba.org>
Autobuild-Date(master): Fri Dec 15 12:09:21 UTC 2023 on atb-devel-224
(cherry picked from commit 4b7329f158)
Autobuild-User(v4-18-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-18-test): Sat Dec 16 15:26:50 UTC 2023 on atb-devel-224
The one case that is no longer handled specially is when the
destination address is IPv4 loopback. This may previously have been
used to avoid flooding the logs when testing. However, that seems
unnecessary - if testing with 127.0.0.1 then make it a public address.
Modernise debug while touching the code.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523
Signed-off-by: Martin Schwenke <mschwenke@ddn.com>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 8fc3872557)
If a client connects to a non-public address first followed by a connect
to public address with the same client_guid and a connection to
the non-public address gets disconnected first, we hit by a use-after-free
talloc_get_type_abort() called from release_ip() as
"xconn" is already gone, taking smbd_release_ip_state with it.
We need to decide between calling ctdbd_unregister_ips() by default, as
it means the tcp connection is really gone and ctdb needs to remove the
'tickle' information. But when a connection was passed to a different
smbd process, we need to use ctdbd_passed_ips() as the tcp connection is
still alive and the 'tickle' information should not be removed within
ctdb.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Volker Lendecke <vl@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit ddf47e7fe3)
This demonstrates the crash that happens if a client connects to a
non-public address first followed by a connect
to public address with the same client_guid and a connection to
the non-public address gets disconnected first, we hit by a
use-after-free talloc_get_type_abort() called from release_ip() as
"xconn" is already gone, taking smbd_release_ip_state with it.
Note that we also need to mark some subtests as flapping
as there's a 2nd problem that happens in the interaction
between smbd processes and ctdb when passing a multichannel
connection to an existing process, it means we sometimes
loose the 'tickle' information within ctdb to that tcp connection.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit 082c7df4d0)
This is similar to ctdbd_unregister_ips(), but with the
difference that ctdb keeps the 'tickle' information for
the tcp connection alive, because another smbd process
took care of that tcp connection in a multichannel scenario.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit 2e784789d7)
This reverts the effect of ctdbd_register_ips().
We'll use this in order to disconnect individual
multichannel connections.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit f3a03f3f77)
This is to remove a callback during rundown of smbds.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Volker Lendecke <vl@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit 75aa669394)
We do the dispatching to multiple handlers in ctdbd_msg_call_back()
and we don't need more than one message from ctdb.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit 77a559432f)
We dump the structure into the socket, so we need to zero the content
including possible padding.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit 240139370a)
With multichannel a tcp connection is registered first with
a temporary smbd process, that calls CTDB_CONTROL_TCP_CLIENT
first and then passes the tcp connection to the longterm smbd
that already handles all connections belonging to the specific
client_guid. That smbd process calls CTDB_CONTROL_TCP_CLIENT
again, but the 'tickle' information is already there.
When the temporary smbd process exists/disconnects from ctdb
or calls CTDB_CONTROL_TCP_CLIENT_DISCONNECTED, the 'tickle'
information is removed, while the longterm smbd process
still serves the tcp connection.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit 037e8e449d)
With multichannel a ctdb connection from smbd may hold multiple
tcp connections, which can be disconnected before the smbd
process terminates the whole ctdb connection, so we a
way to remove undo 'CTDB_CONTROL_TCP_CLIENT' again.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15523
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit c6602b686b)
We could also remove the src_addr and dest_addr helper variables
completely, but that would be too much for this commit.
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Martin Schwenke <martin@meltin.net>
(cherry picked from commit 5f52d140f7)
Align getxattr logic with the rest of xattr hooks: call ceph_fgetxattr
with appropriate io-fd when 'is_pathref' is false; otherwise, call
ceph_getxattr.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15440
Signed-off-by: Shachar Sharon <ssharon@redhat.com>
Reviewed-by: Anoop C S <anoopcs@redhat.com>
Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Günther Deschner <gd@samba.org>
Autobuild-Date(master): Thu Nov 30 12:32:29 UTC 2023 on atb-devel-224
(cherry picked from commit 83edfcff5c)
Autobuild-User(v4-18-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-18-test): Mon Dec 11 13:21:02 UTC 2023 on atb-devel-224
BUG: https://bugzilla.samba.org/show_bug.cgi?id=13577
Signed-off-by: Samuel Cabrero <scabrero@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Tue Sep 5 22:11:46 UTC 2023 on atb-devel-224
(cherry picked from commit f3c632e74b)
Autobuild-User(v4-18-test): Jule Anger <janger@samba.org>
Autobuild-Date(v4-18-test): Wed Nov 29 15:55:45 UTC 2023 on atb-devel-224
